IMAGING NEUROINFLAMMATION IN BIPOLAR DISORDER WITH RADIOLABELED ARACHIDONIC ACID: We reported that brain uptake of radiolabeled arachidonic acid (1-14CAA) could be used to assess neuroinflammation in different animal models, and confirmed using PET and the positron-emitting isotope 1-11CAA the presence of upregulated AA incorporation as a marker of neuroinflammation in Alzheimer disease (AD) patients (Esposito et al., J Nucl Medicine. 2008 49:1414-21). Based on this work, in collaboration with researchers at the at Weill Cornell Medical College and the New York Psychiatric Institute, we are conducting a NIH-grant supported protocol to extend this observation and to neuroimage neuroinflammation with 1-11CAA using PET in patients with bipolar disorder. This research is based in part on our report that the postmortem brain from bipolar disorder patients demonstrated increased markers of neuroinflammation, in association with increased markers of upregulated arachidonic acid (AA) metabolism. Taken in the context of our findings that mood stabilizers used in BD downregulate rat brain AA metabolism, Dr. Elizabeth Sublette at New York Psychiatric Institute has initiated a NIH Grant-supported collaborative clinical protocol to image brain AA metabolism, using our PET method in depressed patients with BD, compared to healthy controls. TEST-RETEST PROPERTIES OF 11CARACHIDONIC ACID BRAIN INCORPORATION MEASUREMENTS, We reported that brain uptake of radiolabeled arachidonic acid could be used to assess neuroinflammation in different animal models, and confirmed using PET and the positron-emitting isotope 11CAA upregulated AA incorporation as a marker of neuroinflammation in Alzheimer disease (AD) patients (Esposito et al., J Nucl Medicine. 2008 49:1414-21). However, although regional incorporation coefficient (K*) of arachidonic acid (AA) into membrane phospholipids of the human brain can be quantified using PET and 11CAA, reproducibility of 11CAA measurements remains undetermined in test-retest paradigms. In collaboration with Dr. Sublette and colleagues, under an IRB approved protocol, we assessed test-retest properties of K*, and investigated the effects on K* estimation and reproducibility of using an image-derived input function (IDIF) and a population-based metabolite correction (PBMC), instead of blood sampling and individual metabolite assays. Eleven healthy volunteers underwent a dynamic PET scan with 11CAA; five had a second scan about 6 weeks later to mimic a pre- and post-treatment study design. In all scans, arterial blood samples were collected to measure 11CAA radioactivity in arterial plasma, and an IDIF was calculated. In 5 scans, the plasma fraction of unchanged 11CAA was determined. K* was estimated using the model we proposed in the literature for 11CAA, with both arterial blood data and IDIF as input function, corrected for the fraction of unchanged 11CAA using both PBMC with published values and individually measured values (when available). Time-stability of K* was assessed, and reproducibility calculated in the test-retest subset using the percent difference. The average percent difference values obtained for K* using either arterial blood data or IDIF, assuming the same PBMC for both, were in the range 6.7% to 13.9%, were not statistically different, from each other. Results show that one can simplify acquisition of data for K* for 11CAA by eliminating arterial blood sampling and metabolite analysis and shorten scan duration to 33 minutes, and retain reliability and validity comparable to results from a 60 minute scan with an arterial input function. These results are being prepared for publication. IMAGING FAILING SYNAPTIC INTEGRITY WITH PROGRESSIVE BRAIN DISEASE. We published blood flow changes in the human brain, using 15O-H20 PET, in response to visual flash stimulation at different frequencies. We showed progressive synaptic failure in Alzheimer disease patients in relation to dementia severity (Mentis et al. Am J Psychiatry 155, 785, 1998). We are planning a collaboration with Professor Elliot Hong, at the University of Maryland School of Medicine, to map brain regional evoked responses (event related potentials) using electroencephalography, in response to the same visual flash stimulation paradigm, in control and schizophrenic patients, as a potential marker of synaptic changes in schizophrenia. Synaptic changes may underlie cognitive deficits, and might be identified using ERP and our flash stimulation paradigm. We plan to extend the method using fMRI in a later collaboration.